Bank / Analdea

Estimated: these come from the final predictions made before each bus arrived, not from where the bus was seen.

What it measures

How often buses were on time at this stop in each hour of the day. Two rules are shown side by side. The OC Transpo rule allows up to one minute early and five minutes late, and is counted at timepoints only. The other counts a bus within five minutes either way of the times riders see, at every stop. The bars use the second.

• MeasuredWe saw it happen.
• EstimatedModelled, shown with a range.
• ExplainedHow it is worked out, in words.

Where it comes from

• EstimatedSettled final predictions for bus arrivals at this stop
• MeasuredPublished timetables, dated

How we work it out

1. We take each bus arrival at this stop over the last four weeks, with the final prediction made just before it.
2. We group arrivals by the hour they were scheduled, weekdays apart from weekends.
3. On the riders’ rule an arrival counts when it is within five minutes either way of the posted time. On the OC Transpo rule it counts only at timepoints, up to one minute early and five minutes late.
4. An hour with too few arrivals shows no number and says so. It is never shown as zero.

Worked example

How sure we are

These come from final predictions, not from where each bus was seen, so they can differ from the headline figure, which is judged from bus positions. The page shows how many buses each hour holds.

Limits, and what would change it

• The two columns can cover different arrivals, so they are not one rule measured twice.
• Early and late count alike on the riders’ rule; on the OC Transpo rule early is stricter.
• An hour with few buses can swing a lot.